US 12,340,176 B2
Techniques for verifying veracity of machine learning outputs
Alexander C. Magary, Contoocook, NH (US); and Kareem Serageldin, London (GB)
Assigned to AMKS Investments I LLC, Contoocook, NH (US)
Filed by AMKS Investments I LLC, Contoocook, NH (US)
Filed on Feb. 23, 2024, as Appl. No. 18/586,179.
Application 18/586,179 is a continuation of application No. 18/486,023, filed on Oct. 12, 2023, granted, now 11,966,704.
Prior Publication US 2025/0124230 A1, Apr. 17, 2025
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 17/00 (2019.01); G06F 40/30 (2020.01)
CPC G06F 40/30 (2020.01) 32 Claims
OG exemplary drawing
 
1. A system for verifying information comprising:
at least one computer hardware processor; and
at least one computer-readable storage medium storing processor-executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a method comprising:
(A) accessing an input from the at least one computer-readable storage medium or received via at least one communication network, the input comprising one or more verifiable statements;
(B) verifying, using a first trained machine learning (ML) model and first reference data stored in at least one first datastore, the one or more verifiable statements to produce ML outputs, the ML outputs comprising first verification results indicating which of the one or more verifiable statements has been verified, the first trained ML model trained for natural language comprehension and configured to process at least one verifiable statement and at least a portion of reference data as input to generate output comprising verification results;
(C) when it is determined that at least one of the one or more verifiable statements remains unverified based on the first verification results,
accessing at least one second datastore having second reference data attesting to veracity of the input; and
verifying, using (i) the first trained ML model or a second trained ML model and (ii) the second reference data accessed from the at least one second datastore, the at least one unverified statement to produce second verification results; and
(D) displaying, via a graphical user interface, output indicating whether one or more of the one or more verifiable statements have been verified based on at least one of the first verification results or the second verification results.